readings covered human perception foveal vision
play

Readings Covered Human Perception Foveal Vision thumbnail at arms - PowerPoint PPT Presentation

Readings Covered Human Perception Foveal Vision thumbnail at arms length Lecture 5: Perception sensors/transducers Information Visualization Ware, Chapter 5: Visual Attention and Information That Pops Out psychophysics: determine


  1. Readings Covered Human Perception Foveal Vision ◮ thumbnail at arm’s length Lecture 5: Perception ◮ sensors/transducers Information Visualization Ware, Chapter 5: Visual Attention and Information That Pops Out ◮ psychophysics: determine characteristics CPSC 533C, Fall 2006 Ware, Chapter 6: Static and Moving Patterns ◮ relative judgements: strong The Psychophysics of Sensory Function, S. S. Stevens, Sensory ◮ absolute judgements: weak Communication, MIT Press, 1961, pp 1-33. ◮ continuing theme Tamara Munzner Graphical Perception: Theory, Experimentation and the Application to ◮ different optimizations than most machines the Development of Graphical Models William S. Cleveland, Robert UBC Computer Science ◮ eyes are not cameras McGill, J. Am. Stat. Assoc. 79:387, pp. 531-554, 1984. ◮ perceptual dimensions not nD array 25 September 2006 ◮ (brains are not hard disks) Foveal Vision Equal Legibility Foveal Touch Eyes ◮ thumbnail at arm’s length ◮ saccades [video] ◮ small high resolution area on retina ◮ star-nosed mole ◮ fovea: high-resolution samples ◮ if fixated on center point ◮ brain makes collage ◮ vision perceived as entire simultaneous field ◮ fixation points: dwell 200-600ms ◮ moving: 20-100ms [www.cs.nyu.edu/ ∼ yap/visual/home/proj/foveation.html] [psy.ucsd.edu/ sanstis/SABlur.html] [www.nature.com/nsu/010329/010329-6.html] [brain.nips.ac.jp/event/work131030/Catania and Kaas, 1997.pdf] [vision.arc.nasa.gov/personnel/jbm/home/projects/osa98/osa98.html/ [svi.cps.utexas.edu/examples foveated.htm] Ears Other Modalities Psychophysical Measurement Nonlinear Perception of Magnitudes sensory modalities not equally discriminable p ◮ perceived as temporal stream Stevens’ Power Law: I = S ◮ but also samples over time ◮ barrier: lack of record/display technology ◮ JND: just noticeable difference Shock Heaviness Taste Length ◮ hard to filter out when not important ◮ haptics maturing ◮ visual vs auditory attention ◮ increment where human detects change ◮ ”haptic visualization” very new ◮ implications Area ◮ average to create “subjective” scale ◮ smell, taste Brightness ◮ harder to create overview? Sensation ◮ out-there SIGGRAPH ETech demos ◮ low-level perception more uniform than Volume ◮ hard to use as separable dimension? ◮ characterization possible after technology Loudness ◮ ’sonification’ still very niche area high-level cognition across subjects barriers fall Smell ◮ alternative: supporting sound enhances immersion Intensity [Stevens, On the Theory of Scales of Measurement, Science 103:2684, 1946] Dimensional Dynamic Range Dimensional Ranking: Accuracy Cleveland vs. Mackinlay: Quantitative Weber’s Law ◮ linewidth: limited discriminability Mackinlay Cleveland ◮ spatial position best for all types ◮ ratio of increment threshold to background intensity is constant position position along common scale ◮ relative judgements within modality Quantitative Ordinal Nominal position along nonaligned scales ∆ I length length, direction, angle Position Position Position = K angle Length Density Hue I Angle Saturation Texture slope ◮ Cleveland example: frame increases Hue Connection Slope area area Area Texture Containment volume, curvature accuracy volume Volume Connection Density Density Containment Saturation density shading, color saturation Saturation Length Shape saturation Hue Angle Length hue Texture Slope Angle Connection Area Slope texture Containment Volume Area connection Graphical Perception: Theory, Experimentation and the Application to the Development Shape Shape Volume containment of Graphical Models. William S. Cleveland, Robert McGill, J. Am. Stat. Assoc. 79:387, [Mackinlay, Automating the Design of Graphical Presentations of Relational pp. 531-554, 1984. shape Information, ACM TOG 5:2, 1986] [mappa.mundi.net/maps/maps 014/telegeography.html]

  2. Cleveland Suggestions Preattentive Visual Dimensions Many Preattentive Visual Dimensions Not All Dimensions Preattentive hue ◮ color (hue) alone: preattentive shape ◮ attentional system not invoked texture ◮ search speed independent of distractor count length parallelism ◮ dot chart over pie or bars width size ◮ direct differences over superimposed orientation curves curvature intersection ◮ framed rectangles over shading on maps intensity [www.csc.ncsu.edu/faculty/healey/PP/PP .html] flicker ◮ demo direction of motion stereoscopic depth [Chris Healey, Preattentive Processing, light direction, . . . www.csc.ncsu.edu/faculty/healey/PP/PP .html] [www.csc.ncsu.edu/faculty/healey/PP/PP .html] Preattentive Visual Dimensions Preattentive Visual Dimensions Separable vs. Integral Dimensions Glyphs ◮ color alone: preattentive ◮ color alone: preattentive ◮ shape alone: preattentive ◮ shape alone: preattentive ◮ not all dimensions separable ◮ composite graphical mark ◮ encoding using multiple dimensions ◮ large-scale individual glyphs vs. small-scale texture fields ◮ combined hue and shape (demo) ◮ combined hue and shape (demo) ◮ grouping into large-scale patterns ◮ requires attention ◮ search speed linear with distractor count ◮ integral vs. separable analysis ◮ when do they help? color color color size x-size red-green location motion shape orientation y-size yellow-blue [Colin Ware, Information Visualization: Perception for Design. Morgan Kaufmann 1999.] [www.csc.ncsu.edu/faculty/healey/PP/PP .html] [www.csc.ncsu.edu/faculty/healey/PP/PP .html] Glyphs: InfoBug Glyphs: InfoBug Small Multiples Array Glyphs: Bray Gestalt Laws ◮ Web sites circa 1996 ◮ software management ◮ # pages: base diameter ◮ # outlinks: globe diameter ◮ # inlinks: height ◮ principles of pattern perception ◮ domain: hue ◮ ”gestalt”: German for ”pattern” ◮ original proposed mechanisms wrong ◮ rules themselves still useful ◮ Pragnatz ◮ simplest possibility wins [Information Rich Glyphs for Software Management, IEEE CG&A 18:4 1998, [Information Rich Glyphs for Software Management, IEEE CG&A 18:4 1998, Bray, Measuring the Web, WWW5, 1996. www.cs.cmu.edu/ ∼ sage/Papers/CGAglyph/CGAglyph.pdf] www5conf.inria.fr/fich html/papers/P9/Overview.html www.cs.cmu.edu/ ∼ sage/Papers/CGAglyph/CGAglyph.pdf] Gestalt Principles Proximity Similarity Continuity ◮ smooth not abrupt change ◮ overrules proximity ◮ proximity, similarity, continuity/connectedness/good continuation ◮ closure, symmetry ◮ common fate (things moving together) ◮ figure/ground, relative sizes [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000] [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000] [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000]

  3. Connectedness Closure Symmetry Common Fate ◮ can overrule size, shape ◮ overrules proximity ◮ demo ◮ emphasizes relationships ◮ tepserver.ucsd.edu/ ∼ jlevin/gp/time-example- common-fate [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000] [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000] [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000] [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000] Relative Size Figure/Ground Graph Drawing Tension Motion ◮ smaller components perceived as objects ◮ works for preattentive/grouping ◮ determined by combination of previous laws ◮ node placement ◮ less studied than static dimensions ◮ close ◮ Michotte on causality ◮ newer infovis/motion work by Lyn Bartram ◮ proximity ◮ far ◮ biological motion ◮ visual popout of long edge ◮ demo ◮ either ◮ connectedness ◮ tradeoffs abound in infovis! [www.research.att.com/sw/tools/graphviz] ◮ grammars ◮ node-link graphs ◮ maps [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000] [Information Visualization: Perception for Design. Ware, Morgan Kaufmann, 2000] [www.psy.vanderbilt.edu/faculty/blake/biowalker.gif] More Perception Presentation Topic Choices ◮ Rensink grad course taught every few years ◮ Perceptual Issues in Visual Interface Design, CPSC 532E Jan 2003 http://www.cs.ubc.ca/ ∼ rensink/courses/cpsc532E/ ◮ Special Topics in Perception: Visual Display Design, PSYCH 579 Jan 2006 http://www.psych.ubc.ca/ ∼ rensink/courses/psyc579/

Recommend


More recommend